MARINE ECOLOGY PROGRESS SERIES

Mar Ecol Prog Ser

Vol. 321: 167181, 2006

Published September 8

Trophic transfer of trace metals from the polychaete

worm Nereis diversicolor to the polychaete N. virens
and the decapod crustacean Palaemonetes varians
P. S. Rainbow1,*, L. Poirier2, B. D. Smith1, K. V. Brix3, 4, S. N. Luoma1, 5
1

Department of Zoology, The Natural History Museum, Cromwell Road, London SW7 5BD, UK

Groupe SMAB, UPRES EA 2160, Laboratoire dEcotoxicologie, Universit de Nantes, Ple Mer et Littoral - BP 92208,
44322 Nantes Cedex 3, France
3

EcoTox, 8963 Hampton Landing Drive E, Jacksonville, Florida 32256, USA

Rosenstiel School of Marine and Atmospheric Sciences, University of Miami, Miami, Florida 33149, USA

United States Geological Survey, 345 Middlefield Road, Mail Stop 465, Menlo Park, California 94025, USA

ABSTRACT: Diet is an important exposure route for the uptake of trace metals by aquatic invertebrates, with trace metal trophic transfer depending on 2 stages assimilation and subsequent accumulation by the predator. This study investigated the trophic transfer of trace metals from the sediment-dwelling polychaete worm Nereis diversicolor from metal-rich estuarine sediments in
southwestern UK to 2 predators another polychaete N. virens (Cu, Zn, Pb, Cd, Fe) and the decapod
crustacean Palaemonetes varians (Cu, Zn, Pb, Cd, Fe, Ag, As, Mn). N. virens showed net accumulation of Cu, Zn, Pb and Cd from the prey; accumulation increased with increasing prey concentration,
but a coefficient of trophic transfer decreased with increasing prey concentration, probably because
a higher proportion of accumulated metal in the prey is bound in less trophically available (insoluble)
detoxified forms. The trace metal accumulation patterns of P. varians apparently restricted significant
net accumulation of metals from the diet of N. diversicolor to just Cd. There was significant mortality
of the decapods fed on the diets of metal-rich worms. Metal-rich invertebrates that have accumulated
metals from the rich historical store in the sediments of particular SW England estuaries can potentially pass these metals along food chains, with accumulation and total food chain transfer depending
on the metal assimilation efficiencies and accumulation patterns of the animal at each trophic level.
This trophic transfer may be significant enough to have ecotoxicological effects.
KEY WORDS: Trace metals Trophic availability Assimilation efficiency Accumulation pattern
Nereis Palaemonetes
Resale or republication not permitted without written consent of the publisher

Benthos living in contaminated sediments could be

an important vector for moving potentially toxic metals
into aquatic food webs. It is becoming increasingly
apparent that the diet is an important route for the
uptake of trace metals by predators (Wang 2002), but
the general principles controlling trace metal trophic
transfer are not fully known (Reinfelder & Fisher 1991,
Wang & Fisher 1999, Wang 2002, Wallace & Luoma
2003).

The use of radioactive tracers has allowed the development of standard techniques for the measurement of
trace metal assimilation efficiencies (Wang & Fisher
1999), with an associated expansion of such studies in
a variety of invertebrates including bivalves (Wang &
Fisher 1996, Reinfelder et al. 1997, Chong & Wang
2000, Ng et al. 2005), gastropods (Cheung & Wang
2005), barnacles (Wang & Rainbow 2000, Rainbow et
al. 2003), copepods (Xu & Wang 2002, 2004) and decapod crustaceans (Wallace & Luoma 2003, Rainbow et
al. 2006). Assimilation efficiency (AE) is a measure of

*Email: p.rainbow@nhm.ac.uk

Inter-Research 2006 www.int-res.com

INTRODUCTION

168

Mar Ecol Prog Ser 321: 167181, 2006

bioavailability from diet, characterised by the efficiency of release of trace metals from ingested food in
the gut and uptake into the animal tissues, initially the
epithelial lining of the digestive tract. The accumulation of trace metals that have been assimilated then
depends on the accumulation pattern of the ingesting
animal (Rainbow 2002). In some invertebrates all metal
taken up is accumulated in the animal with negligible
excretion, while in others most or all of the assimilated
metal may be excreted without longer-term accumulation (Rainbow 2002, Luoma & Rainbow 2005). Thus, in
nature, trophic transfer may lead to strong or negligible
net accumulation of trace metals even if bioavailability
(assimilation efficiency) is high.
An alternative approach to assessing the outcome of
trophic transfer is to use longer-term field or laboratory
exposures and measure changes in net accumulated
concentrations of predators feeding on different prey.
Thus, Blackmore (2000) showed that the predatory
gastropod Thais clavigera from adjacent coastal sites
in Hong Kong had different accumulated concentrations of Zn and Cu, according to whether the predominant diet item was the gastropod Siphonaria japonica
(raised dietary availability of Cu) or the barnacle
Balanus amphitrite (raised dietary availability of Zn),
as confirmed in a long-term laboratory experiment:
Blackmore & Morton 2002). Similarly, Rainbow et al.
(2004) showed that the atypically high accumulated Cu
concentrations in the polychaete worm Nereis diversicolor from Restronguet Creek, UK, an estuary with
very high sediment concentrations of trace metals, is
trophically available to the predaceous polychaete
worm N. virens, and brings about high accumulated
copper concentrations in the latter. The study of the
trophic transfer of copper requires stable metal analyses, given the absence of suitable Cu radiotracers for
assimilation efficiency and biokinetic experiments.
The development of Cu stable isotope techniques will
fill this technological gap (Croteau et al. 2004), but in
the meantime the analyses of total Cu levels in prey
and predator provide the best insight into the trophic
transfer of this essential but relatively toxic trace metal.
The present study used laboratory exposures to
address 2 specific questions with regard to the trophic
transfer of trace metals along food chains. Do the polychaete predator Nereis virens, and the decapod crustacean predator Palaemonetes varians, accumulate
trace metals from N. diversicolor as prey? If so, does
the rate of metal accumulation differ among prey
whose exposure histories differed over their lifetimes
(resulting in different accumulated trace metal concentrations)? An implication of the latter question is
that the subcellular fractionation of accumulated metals in the prey may also vary with accumulated concentration and such variation may result in differential

assimilation and subsequent trophic transfer (Wallace

& Lopez 1996, 1997, Wallace & Luoma 2003, Rainbow
et al. 2006).
The study is an extension of the earlier work on the
trophic transfer of copper from Nereis diversicolor to
N. virens (Rainbow et al. 2004) and is complementary
to a radiotracer study of the Zn and Cd assimilation
efficiencies of the decapod crustacean Palaemonetes
varians feeding on N. diversicolor. In all cases we used,
for prey, individuals from metal-contaminated estuaries with different histories of contamination (Rainbow
et al. 2006), thereby varying the dietary exposure of
the predator. This approach obviates the need to assume that a spike of radioactively labelled metal is uniformly distributed among the subcellular fractions of
accumulated metal that result from a lifetime of exposure (e.g. Rainbow et al. 2004, Wallace & Lopez 1996).
The counties of Devon and Cornwall in SW England
have many estuaries with metal-rich sediments,
derived from mining activities from (particularly) the
nineteenth century, when the area boasted more than
1000 mines and was one of the most important metalproducing regions in the world. The sulphide-rich sediments of these estuaries contain arsenic, cadmium,
chromium, copper, iron, lead, manganese, mercury,
nickel, silver, tin, tungsten and zinc, in different proportions and significance according to the original ore
mined (Dines 1969), and the estuarine biota have differentially raised concentrations of accumulated trace
metals (Bryan et al. 1980, 1985, Bryan & Gibbs 1983).
Trace metal concentrations in sediment and the infaunal polychaete Nereis diversicolor collected from
Restronguet Creek in 2000 matched those measured
20 yr previously (Berthet et al. 2003). We have found a
similar lack of change in metal concentrations of sediments, N. diversicolor and the infaunal bivalve Scrobicularia plana over 20 to 30 yr in Restronguet Creek,
as well as in the Gannel, Tavy and East and West Looe
estuaries (S. N. Luoma et al. unpubl.), showing that
these southwest England estuarine sediments have
maintained their high metal concentrations and
bioavailabilities over decades.
Prey worms were collected from 5 metal-rich Devon
and Cornwall estuaries (Restronguet Creek, Gannel,
Tavy, East Looe and West Looe; see Fig. 1), and from 2
uncontaminated estuaries, the Dart in Devon (Fig. 1)
and the Blackwater in Essex in east England (not
shown in Fig. 1). We expected the prey worms to contain differentially raised concentrations of accumulated trace metals (Bryan & Hummerstone 1971, Bryan
et al. 1980, Bryan & Gibbs 1983), with potentially different internal distributions of accumulated metals into
soluble and insoluble detoxification pathways with
probable consequences for their trophic bioavailabilities (Wallace & Luoma 2003, Rainbow et al. 2006).

Rainbow et al.: Trophic transfer of trace metals

Of the estuaries chosen, Restronguet Creek is a

branch of the Fal estuary system, receiving discharge
from the Carnon River, draining a region with a long
history of mining for metals (Dines 1969, Bryan &
Gibbs 1983). Restronguet Creek correspondingly contains extraordinarily high levels of As, Cu, Fe, Mn and
Zn (Bryan et al. 1980, Bryan & Gibbs 1983, Berthet et
al. 2003). The presence in Restronguet Creek of populations of invertebrates with tolerance to copper and
zinc (Bryan & Hummerstone 1971, Bryan & Gibbs
1983, Mouneyrac et al. 2003) confirms that the local
bioavailabilities of these 2 metals are ecotoxicologically significant. The Gannel estuary has been
reported to have high bioavailabilities of Pb and Zn,
the Tavy of Cu, the East Looe of Ag, Cu and Pb, and
the West Looe of Cu and Pb, with no raised trace metal
bioavailabilities reported for the Dart (Bryan et al.
1980).
Two predators were the same as used (separately) in
the 2 earlier studies (Rainbow et al. 2004, 2006). Palaemonetes varians is a willing predator and will quickly
eat a meal of Nereis diversicolor offered. AE is measured in radiotracer experiments and is a parameter
that is independent of subsequent physiological handling of accumulated metals after assimilation. P. varians is an ideal model organism therefore for AE radiolabelling experiments and associated investigation of
the principles controlling trace metal assimilation
(Rainbow et al. 2006). Palaemonid decapod crustaceans,
such as P. varians, however, regulate body concentrations of zinc by matching the rate of zinc excretion to
that of zinc uptake over a wide range of dissolved Zn
bioavailabilities (White & Rainbow 1982,1984a,b, Amiard et al. 1987, Nugegoda & Rainbow 1989, Rainbow
1998). Thus, palaemonid decapods do not accumulate
extra body zinc when exposed to high dissolved zinc
bioavailabilities even if as much as 13% of the total Zn
body burden is turned over per day (White & Rainbow
1984a, Rainbow 1998). The constancy of intraspecific
body Zn concentrations between specimens collected
from many different sites indicates that Zn taken up
from the diet is similarly excreted to regulate the accumulated body concentration (Bryan 1968, White &
Rainbow 1982, Rainbow 1998). Although difficult to
verify in the absence of suitable radiotracers, it appears
from field measurements and dissolved stable metal
exposure experiments that palaemonids also regulate
body concentrations of Cu (Bryan 1968, White & Rainbow 1982, Amiard et al. 1987, Rainbow 1998). The regulation of body trace metal concentrations by palaemonid decapod crustaceans may also extend to other
essential metals but not to non-essential metals such as
Cd and Pb (White & Rainbow 1982, 1986, Amiard et al.
1987, Rainbow 1998). Thus, it is unlikely that P. varians
will show extra accumulation of zinc and copper with

169

increasing uptake from the diet, unless the uptake rate

is so high as to cause regulation breakdown (Rainbow
1998). On the other hand, there is the likelihood that P.
varians will show differential accumulation of nonessential metals (e.g. Ag, Cd, Pb) given sufficient differential metal uptake from the diet.
We therefore needed a predator that would show net
accumulation of 1 or more essential metals such as Cu
or Zn, as well as non-essential metals (e.g. Cd, Pb). The
choice fell on another Nereis species, N. virens. N.
diversicolor is a net accumulator of copper (Bryan &
Hummerstone 1971, Bryan et al. 1980, Amiard et al.
1987, Zhou et al. 2003, Geffard et al. in press), causing
Rainbow et al. (2004) to conclude (and indeed subsequently confirm) that N. virens (a predator able to feed
on N. diversicolor) might accumulate copper, specifically from a diet of its congener. Thus, N. virens could
be expected to show differential accumulation of copper in this study given differential uptake of copper
from the diet. On the other hand, N. diversicolor
appears to regulate the body concentration of zinc,
with implications therefore for possible regulation of
body zinc concentration by N. virens. Thus, field concentrations of N. diversicolor vary little (Bryan & Hummerstone 1973a, Bryan et al. 1980), and Amiard et al.
(1987) provided experimental evidence that increased
dissolved availabilities of zinc caused no increase in
accumulated body zinc concentrations. It remained
likely therefore that N. virens would be able to regulate body zinc concentrations (at least to some extent)
across a range of dietary zinc uptake rates. There is
evidence from laboratory experiments (Amiard et al.
1987, Geffard et al. 2005) and the field (Bryan & Hummerstone 1971, 1973a,b) that N. diversicolor is a net
accumulator of the non-essential metals Cd and Pb,
and also for the essential metal manganese. Again it
was likely that N. virens would be a net accumulator of
these metals as well.

MATERIALS AND METHODS

Collection. Nereis diversicolor were collected by
hand from intertidal mudflats (upper 20 cm sediment
depth) in the estuaries of the Dart, East Looe, Gannel,
Restronguet Creek, Tavy and West Looe in southwest
England (Fig. 1) between 1 and 4 September 2003, and
from the Blackwater estuary (eastern England) on
16 September 2003 (collection details in Table 1).
They were transported back to the laboratory in cool
boxes in sediment from the collection site. The worms
were then kept in sediment from the site of origin
covered by artificial seawater (TM: Tropic Marin,
Tropicarium Buchschlag, Dreieich, Germany) at a
salinity of 16 at 15C.

170

Mar Ecol Prog Ser 321: 167181, 2006

Table 1. Collection details of the polychaete worm Nereis

diversicolor from UK estuaries

Fig. 1. Locations of the 6 collection sites in SW England

Nereis virens were purchased from SEABAIT,

where they are reared commercially, and held in TM at
a salinity of 33 at 15C. Palaemonetes varians were collected from salt marsh pools at Tollesbury, Blackwater
estuary (51 45 59 N, 0 50 08 E) on 16 September
2003, and transported to the laboratory in cool boxes.
The decapods were kept in aquaria in artificial seawater (TM) at a salinity of 6 at 15C. Adult P. varians of
approximately equal size (about 3.5 cm total length)
were used for experiments.
Feeding experiments. We used Nereis virens as
predator in 2 experiments; Palaemonetes varians as
predator in a third experiment.
Food: Nereis diversicolor from each collection site
were placed in TM at 16 salinity at 15C for 2 d without

Site

Co-ordinates

Collection date
(2003)

Restronguet Creek
East Looe
Gannel
West Looe
Tavy
Dart
Blackwater

50 12.36 N, 5 5.41 W
50 22.38 N, 4 27.74 W
50 24.32 N, 5 05.17 W
50 21.82 N, 4 28.86 W
50 27.12 N, 4 10.17 W
50 23.98 N, 3 37.35 W
51 44.08 N, 0 41.34 E

1 Sep
2 Sep
2 Sep
3 Sep
3 Sep
4 Sep
16 Sep0

sediment to allow depuration of the gut before they

were frozen for use as food in feeding experiments or
for metal analysis. Because the jaws of nereid polychaetes may contain high concentrations of metals
(Bryan & Gibbs 1980), only the bodies of decapitated
worms were used as a food source in experiments.
Typically, frozen decapitated worms were cut into
about 5 separate portions, each about 0.5 cm long, and
thawed to be used as discrete food packages (see
below). For each collection site, 5 worm sections were
pooled into each of typically 3 replicates and weighed
to constant dry weight at 60C for estimation of the
average dry weight (Table 2), and consequently trace
metal content, of each food package. For metal analysis, 5 worms from each site were decapitated and the
decapitated bodies frozen for metal analysis. Preliminary analyses had confirmed lack of significant variation between metal concentrations of different sections
of body along the length of the worm behind the head.
Nereis virens as predator (Expt 1), 11 September 2003:
Initial concentrations of metals were determined in 8
N. virens which had been kept individually in 50 ml
TM in acid-washed 100 ml beakers for 24 h (salinity 33,
15C) to depurate all gut contents before being frozen
for metal analysis; 150 further N. virens were placed
individually into acid-washed 100 ml beakers with 50 g
acid-washed sand (Merck) and 50 ml TM at salinity 33
at 15C. These latter worms (n = 25 for each of 6 estuaries) were fed twice daily with 1 thawed section of

Table 2. Nereis diversicolor as food. Mean ( SD) metal concentrations (g g1 dry wt) and dry weights (mg) of body sections of
decapitated frozen worms used as individual meals for N. virens in feeding experiments
Site
Blackwater
Dart
East Looe
Gannel
Restronguet Cr.
Tavy
West Looe

Meal dry wt
2.70.3
2.60.3
2.80.4
1.90.8
2.90.6
2.80.1
2.30.9

Ag

1.570.63
0.580.56
2.280.92
2.060.58
2.240.42
0.540.24
0.660.26

As

Cd

Cu

Fe

Mn

Pb

Zn

8.901.63 0.360.04
6.254.050 35856.8 14.75.32 <1.35<8.52 11322.6
10.30.48 < 0.20< 0.90 9.422.280 51841.9 19.36.23 <1.48<6.82 11421.5
9.742.66 < 0.27<0.61 16.84.410 503 229. 7.302.81 4.381.43 68.810.3
12.23.18 0.330.01
51.723.62 573161. 13.02.60 27.717.8 .14844.9
.11719.5 0.620.18
8582260 15214970. 19.35.90 3.981.90 .20136.3
8.843.76 0.430.10
41.216.40 56993.7 16.93.98 2.120.48 .12521.2
13.22.79 < 0.23< 0.79 26.39.410 427104. 8.361.70 34.515.7 75.37.75

171

Rainbow et al.: Trophic transfer of trace metals

N. diversicolor from 1 of the 6 southwest England estuaries for up to 18 d. Not all worms ate every meal
offered, so each meal taken by the individual worms
was recorded. Thus, the cumulative number of meals
eaten by each worm was known, uneaten meals being
removed. It was intended to sample 8 worms from each
dietary treatment on Days 6, 12 and 18, but the number
actually sampled (5 to 9) depended on the number of
worms regularly feeding. Sampled worms were kept
individually in 50 ml TM in acid-washed 100 ml
beakers for 24 h to depurate all gut contents before
being frozen for metal analysis.
Nereis virens as predator (Expt 2), 7 October 2003:
In a repeat experiment, 10 N. virens were frozen for
initial metal analysis after depuration of gut contents
as described above; 180 further N. virens (n = 30 for
each of 6 estuaries) were separated individually as in
Expt 1, and fed twice daily with 1 thawed section of N.
diversicolor from 1 of 5 southwest England estuaries
(insufficient samples remained from the Dart collection) or from the Blackwater estuary for up to 18 d.
Each meal taken by the individual worms was recorded as in Expt 1; 9 to 10 predatory worms were
sampled from each dietary treatment on Days 6, 12 and
18, and sampled worms were depurated as described
above and frozen for metal analysis.
Palaemonetes varians as predator, 2 October 2003:
We kept 8 P. varians in TM for 24 h (salinity 6, 15C) to
depurate all gut contents before freezing them for initial metal analysis. We placed 4 groups of about 30 (29
to 32) P. varians each in acid-washed aquaria TM at
salinity 6 at 15C. These prawns were fed (twice daily
ad libitum for 2 h during the first 3 d, once daily thereafter), with thawed sections of N. diversicolor from
either the Dart, Gannel, Restronguet Creek or Tavy,
the original intention being to do this for up to 18 d
with sampling at Days 6, 12 and 18. The water in each
tank was changed after each meal. All diet treatments
showed considerable mortality of the prawns, and
samples of prawns were taken on Days 6 (all 4 treatments), 10 (Tavy), 12 (Dart, Gannel, Restronguet
Creek), 13 (Gannel) and 14 (Restronguet Creek) before
provision of that days meal. The choice of day of sampling was made in the light of the mortality of prawns
that had already occurred and the apparent state of
health of the remaining prawns in a treatment. Prawns
were frozen for later analysis of metal contents.
Given the prawn mortalities (unexpected from the
results of the experiments with Nereis virens), it was
decided to run a control experiment in which the
prawns were fed with a diet other than N. diversicolor.
The choice of food in this case was TetraFin flakes
(Tetra), a commercially available fishfood. Seven Palaemonetes varians were frozen as initial samples and
24 prawns kept as described above for up to 18 d

(sampled on Days 6, 12 and 18), fed once a day on

TetraFin. Sampled prawns were treated as described
above.
Metal analysis. Nereis diversicolor (including body
sections), N. virens and Palaemonetes varians that had
been frozen for metal analysis (see foregoing subsections) were subsequently dried to constant weight at
60C and acid-digested in concentrated nitric acid
(Aristar grade, Merck) at 100C. Each digest was made
up to a known volume with double-distilled water and
analysed for some or all of the trace metals Ag, As, Cd,
Cu, Fe, Mn, Pb and Zn by atomic absorption spectrophotometry (AAS) on a Varian SpectrAA 220 FS
spectrophotometer with background correction as
appropriate, or (N. diversicolor body sections only) on
a Vista-Pro CCD Simultaneous ICP-OES (Inductively
Coupled Plasma-Optical Emission Spectrometer). Table
3 shows the results obtained for the standard reference
materials Tort-1 and Tort-2 (lobster hepatopancreas,
National Research Council, Canada), and agreement is
considered satisfactory. All metal concentrations are
quoted in terms of dry weight.
Access to ICP metal analysis was obtained late in the
study. Therefore, analyses of the metal concentrations
in the predator Nereis virens were carried out by AAS,
those on the food source N. diversicolor and the preda-

Table 3. Comparisons of measured and certified concentrations of trace metals (g g1 dry wt 95 % confidence limits)
in lobster hepatopancreas standard reference materials Tort-1
and Tort-2 measured by atomic absorption spectrophotometry
(AAS) and inductively coupled plasma-optical emission spectrometry (ICP-OES). na: not available
Metal

Measured conc.

Certified conc.

AAS
Tort-1
Cd
Cu
Fe
Pb
Zn

32.9 5.8
433 28
271 65
9.5 2.5
190 14

26.3 2.1
439 22
186 11
10.4 2.0
177 10

Tort-2
Cd
Cu
Fe
Zn

26.5 1.4
105 13
115 10
206 27

26.7 0.6
106 10
105 13
180 6

ICP-OES
Tort-2
Ag
As
Cd
Cu
Fe
Mn
Pb
Zn

2.4 1.2
22.3 0.5
25.6 0.6
97 13
94 8
11.5 0.4
0.46 0.30
173 4

na
21.6 1.8
26.7 0.6
106 10
105 13
13.6 1.2
0.35 0.13
180 6

172

Mar Ecol Prog Ser 321: 167181, 2006

tor Palaemonetes varians were made by ICP. Inevitably,

such analyses represent a compromise between digest
volumes that are large enough for the analysis of several metals and small enough to obtain metal concentrations above detection limits. Combination of analyses of N. virens from the Expts 1 and 2 gave good data
sets for Cu, Fe and Zn (prey from 7 sites), and partial
but usable data sets for Pb (5 sites) and Cd (4 sites), with
the missing data sets containing an excess of analyses
below detection limits. In the case of P. varians, full data
sets (prey from 4 sites and fishfood control) were obtained for Ag, As, Cd, Cu, Fe, Mn and Zn, with a reduced data set for Pb (prey from 3 sites with reduced
number of analyses within the data sets used).
Statistical analysis. Regression analyses, a priori
analysis of variance (ANOVA) and post hoc a posteriori
ANOVA using Tukeys HSD test for unequal numbers
were carried out using STATISTICA (StatSoft). Percentage data were arcsine-transformed before statistical analysis.

In the case of Cu, there was a significant regression

(Fig. 2, Table 4) for Nereis virens feeding on all N.
diversicolor prey except those from the Blackwater,
the control estuary in Expt 2. There was therefore significant accumulation of Cu from the prey, the largest
increase in the Cu body concentration of the predator

Blackwater

150
100
50
0

Dart

150
100
50
0

East Looe

150

Accumulation of metals from food

As can be seen from Table 2, the prey worms (Nereis
diversicolor) from the different sites represented different accumulated concentrations of trace metals in
the diet of N. virens and Palaemonetes varians.

Nereis virens
Data from the 2 feeding experiments were combined. As the number of meals ingested by each individual Nereis virens was known, it was possible to calculate the total (cumulative) metal ingested by each
worm in the diet and therefore the cumulative amount
of metal ingested per unit weight of predator worm.
The latter was then plotted against the final concentration of accumulated metal in the predator (e.g. Figs. 2,
3 & 4). The slope of each significant best-fit regression
line represents a time-independent trophic transfer
coefficient without units. Figs. 2, 3 & 4 show these plots
for prey from different sites for Cu, Zn and Pb respectively. Table 4 summarises details of the best-fit
regressions for these metals and also for Cd and Fe. It
was also possible to determine mean weight-specific
feeding rates by dividing the cumulative amounts of
metal ingested by the number of days fed. Plots (not
shown) of weight-specific feeding rates against the
concentrations of accumulated metal in the predator
gave very similar results to those presented above
(Figs. 2, 3 & 4; Table 4).

100

Copper conc. (g g1 worm)

RESULTS

50
0

Gannel

150
100
50
0
0

20

40

60

80

100

120

Restronguet Creek
150
100
50
0
0

500

1000

1500

2000

2500

Tavy

150
100
50
0

West Looe

150
100
50
0
0

20

40

60

80

100

120

Copper intake (g g1 worm)

Fig. 2. Nereis virens. Relationships between Cu concentrations of predator worms N. virens and their total Cu intake
from their diet (N. diversicolor from 7 sites) per g weight
predator worm. Best-fit linear regression lines are shown
when significant (regression details in Table 4)

173

Rainbow et al.: Trophic transfer of trace metals

Blackwater

350
300
250
200
150
100
50
0

20
10
0

Gannel

30

Dart

350
300
250
200
150
100
50
0

Zinc conc. (g g1 worm)

East Looe

30

Lead conc. (g g1 worm)

20

East Looe

350
300
250
200
150
100
50
0

10
0

Restronguet Creek

30
20
10
0

Tavy

30
20

Gannel

350
300
250
200
150
100
50
0

10
0

West Looe

30
20
10

Restronguet Creek

350
300
250
200
150
100
50
0

0
0

10

20

30

40

50

60

Lead intake (g g1 worm)

Fig. 4. Nereis virens. Relationships between Pb concentrations of predator worms N. virens and their total Pb intake
from the diet (N. diversicolor from 5 sites) per g weight predator worm. Best-fit linear regression lines are shown when
significant (regression details in Table 4)

Tavy

350
300
250
200
150
100
50
0

West Looe

350
300
250
200
150
100
50
0
0

100

200

300

Zinc intake (g

400

g1

500

600

worm)

Fig. 3. Nereis virens. Relationships between Zn concentrations of predator worms N. virens and their total Zn intake
from the diet (N. diversicolor from 7 sites) per g weight predator worm. Best-fit linear regression lines are shown when
significant (regression details in Table 4).

being derived from the most Cu-rich prey (N. diversicolor from Restronguet Creek) (Fig. 2). Although N.
virens feeding on Restronguet Creek N. diversicolor
accumulated the most copper over the course of the
experiments, the slope of the regression line (defined
as the Cu trophic transfer coefficient) in this case was
the lowest and significantly lower than for the Dart and
East Looe worms (Table 4). Prey from the Dart produced the highest trophic transfer coefficient, significantly higher than the coefficients for prey from West
Looe, Tavy, Gannel and Restronguet Creek (Table 4).
Thus, prey with the lowest Cu concentrations (Table 2)
had either caused no significant Cu accumulation over
the experimental time period (Blackwater) or had the
highest trophic transfer coefficient (Dart), while the
prey with the highest Cu concentration (Restronguet

174

Mar Ecol Prog Ser 321: 167181, 2006

accumulated by the predator in all

cases except for the control Blackwater
worms (Fig. 3, Table 4). There was also
a significant inverse linear relationship
between the trophic transfer coefficient and Zn concentration in Nereis
2
Metal
Slope
Intercept
R
n
p
diversicolor prey (Fig. 5, Table 5).
Site
SE
SE
There are fewer data available for
Copper
lead, given the paucity of lead concenBlackwater
0.328ABC 0.164 14.2 0.75 0.100 38
0.053 ns
trations above detection limits in both
Dart
1.438A 0.575 18.7 3.20 0.269 19
0.023*
prey (Table 2) and predator (Fig. 4).
0.002**
East Looe
0.418AB 0.131 15.3 1.53 0.139 65
BC
Nevertheless, in 4 of the 5 cases for
Gannel
0.113 0.041 16.8 1.08 0.108 64
0.008**
Restronguet Creek 0.039C 0.006 22.1 3.81 0.408 58 < 0.001***
which adequate data are available
Tavy
0.195BC 0.043 14.6 1.23 0.259 61 < 0.001***
(East Looe, Gannel, Restronguet Creek
West Looe
0.215BC 0.102 15.2 1.38 0.068 63
0.039*
and West Looe), there was significant
Zinc
net accumulation by Nereis virens of
108 9.0
0.082 38
0.082 ns
Blackwater
0.194BC 0.108
lead
from the N. diversicolor prey; the
BC
0.023*
Dart
0.355 0.149 0105 10.7 0.160 32
diet of Tavy worms (with the lowest
East Looe
0.360BC 0.163
112 7.5
0.068 69
0.030*
103 7.6
0.405 68 < 0.001***
Gannel
0.700A 0.104
measured Pb concentration: Table 2)
115 6.8
0.092 64
0.015*
Restronguet Creek 0.098C 0.039
did not cause significant net Pb acTavy
0.368B 0.092
112 7.7
0.198 67 < 0.001***
cumulation in the predator (Fig. 4,
AB
100 7.6
0.166 63 < 0.001***
West Looe
0.684 0.196
Table 4). Again, there was a significant
Lead
inverse linear relationship between
East Looe
1.577A 0.365 4.05 0.85 0.368 34 < 0.001***
AB
the trophic transfer coefficient and
Gannel
0.327 0.089 7.72 1.58 0.283 36 < 0.001***
Restronguet Creek 0.733AB 0.314 6.89 1.05 0.142 35 < 0.001***
Pb concentration in the prey (Fig. 5,
0.116 ns
Tavy
1.538AB 0.949 7.78 1.55 0.080 32
Table 5).
West Looe
0.225B 0.054 5.38 1.10 0.338 36 < 0.001***
In the case of the reduced data set
Cadmium
for
cadmium, only 1 set of prey (Black0.037*
Blackwater
1.455A 0.673 0.71 0.18 0.118 37
A
water
Nereis diversicolor) produced
0.756 ns
Gannel
0.240 0.769 1.00 0.14 0.003 41
significant net Cd accumulation in N.
0.647 ns
Restronguet Creek 0.258A 0.560 1.25 0.26 0.213 40
Tavy
0.782A 0.798 1.15 0.28 0.026 38
0.333 ns
virens over the time course of the
Iron
experiment (not illustrated but sum384 26.6 0.032 38
0.286 ns
Blackwater
0.110A 0.102
marised in Table 4). Nevertheless,
271 65.5 0.156 27
0.041*
Dart
0.454A 0.211
even in the absence of a strong gradiA
East Looe
0.006 0.072
384 25.0 0.000 67
0.938 ns
ent of prey Cd concentrations, there
385 19.0 0.067 67
0.589 ns
Gannel
0.037A 0.067
was yet again a significant inverse linRestronguet Creek 0.039A 0.024
382 25.6 0.045 59
0.105 ns
388 25.1 0.027 64
0.198 ns
Tavy
0.084A 0.065
ear relationship between the trophic
339 25.6 0.011 64
0.411 ns
West Looe
0.097A 0.117
transfer coefficient and prey metal
concentration (Fig. 5, Table 5).
Iron differed from the other metals as regards its
Creek) had brought about the lowest trophic transfer
potential net accumulation by Nereis virens from N.
coefficient. Cu trophic transfer coefficients were therefore plotted against prey concentration (Fig. 5,
Table 5). The disproportionately high Cu conTable 5. Nereis virens. Linear regression details for relationships between trophic transfer coefficient of predator worms N. virens (y) and
centration in the Restronguet Creek prey so
trace metal concentration of their diet (N. diversicolor from up to 7 sites)
distorted the scale that a linear regression
per weight of predator worm (g g1) (x). Also shown are regression dewas not significant, while a double log
tails for double-log relationship in the case of copper. ns: not significant
regression (equivalent to a power relationship when not logged) was very highly signifMetal
Slope SE
Intercept SE
R2
n
p
icant (Table 5, Fig. 5). Thus there is a very
significant inverse effect of prey Cu concenCopper
0.0005 0.0003 0.468 0.086 0.118 33
0.051 ns
Zinc
0.018 0.001 0.614 0.100 0.149 33
0.026*
tration on the Cu trophic transfer coefficient
Lead
0.020 0.005 1.057 0.131 0.437 20
0.002**
to the predator.
Cadmium
1.587 0.736 1.486 0.352 0.237 17
0.048*
Data for zinc were treated similarly. Again
Iron
0.0001 0.0001 0.158 0.049 0.037 33
0.286 ns
there was a significant linear relationship beLog copper 0.534 0.067 1.572 0.112 0.670 33 < 0.001***
tween dietary input of metal (Zn) and metal

Table 4. Nereis virens. Regression details for relationships between trace metal
concentration (g g1) of predator worms N. virens (y) and total metal intake
from their diet (N. diversicolor from up to 7 sites) per weight of predator worm
(g g-1) (x). Slopes (trophic transfer coefficients) sharing same index for a particular metal do not differ significantly (p < 0.05). ns: not significant

175

Rainbow et al.: Trophic transfer of trace metals

1.5

0.5

Log coeff.

Copper

1.0
0.5

200

400

600

800

-0.5
-1.0
-1.5
0.5

1000

1.0

1.5

2.0

2.5

3.0

40

50

Log prey conc.

1.0
0.8
0.6

2.5
2
1.5

Zinc

0.4
0.2
0.0
0

50

100

150

200

250

Cadmium
2.0
1.5
1.0
0.5
0.0
0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Trophic-transfer coeff.

Trophic-transfer coeff.

0.0

Log copper

0.0

Lead

1
0.5
0
0

10

20

30

Iron

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
0

500

1000

1500

Fig. 5. Nereis virens. Relationships between trophic transfer coefficient of predator worms N. virens and trace metal concentration of the diet (N. diversicolor from up to 7 sites). Also shown is double-log relationship in case of copper. Best-fit linear
regression lines are shown when significant (regression details in Table 5)

diversicolor. There was only significant net accumulation in 1 case (prey from the Dart) (Table 4), and no significant relationship between any calculated trophic
transfer coefficients (typically themselves not significant models of the data) and prey Fe concentrations
(Fig. 5, Table 5).

Palaemonetes varians
This experiment addressed the net accumulation of
trace metals by the decapod crustacean Palaemonetes
varians from prey, the polychaete Nereis diversicolor
from 4 sites in southwest England, and additionally
from fishfood (TetraFin).
Because it was not possible to estimate the prey
intake of individual decapods, and because decapods
are voracious feeders, without the great variation in
willingness to feed observed for Nereis virens, data
analysis involved the plotting of the accumulated trace
metal concentrations of individual decapods against
number of days feeding, as exemplified in Fig. 6 for Cu
and Fig. 7 for Cd. Best-fit linear regression data for
these plots and for corresponding plots for Zn, Pb, Fe,
Ag, As and Mn are given in Table 6. Only data for surviving decapods are included in the analyses.
As can be seen from Fig. 6 and Table 6, there was no
significant net accumulation of Cu by Palaemonetes

varians feeding on any of the Nereis diversicolor diets.

There was, however, a significant decrease in the
accumulated Cu concentration of P. varians feeding on
fishfood. There was no significant change of weight of
the decapods (positive or negative) feeding on any of
the diets over the period of the experiment (Table 6).
In the case of Zn, there was no significant accumulation by Palaemonetes varians from Nereis diversicolor
from 3 sites, although there was from prey worms from
the Tavy (Table 6). There was a significant decrease in
the accumulated Zn concentration of decapods fed the
fishfood diet (Table 6). There was no significant increase of accumulated Pb concentration in the decapods feeding on N. diversicolor from the Dart, Gannel
or Restronguet Creek (Table 6), the only 3 diets for
which sufficient measurable data are available.
For Cd, on the other hand, over the course of the
experiment, there was significant accumulation of Cd
by the decapods from Nereis diversicolor for 3 out of 4
sites (Fig. 7, Table 6). Dietary concentrations of Cd
were only available for 3 of these 4 diets (Table 2).
Although there was a positive regression between the
slopes of the plots in Fig. 7 and the 3 N. diversicolor diet concentrations measured, this regression
(y = 0.087x 0.019, r2 = 0.830) was not significant.
There was no significant accumulation of Fe, Ag or
As by Palaemonetes varians from Nereis diversicolor
from any of the 4 sites (Table 6). There was a signifi-

176

Mar Ecol Prog Ser 321: 167181, 2006

Dart
300

1.0

200

0.5

Dart

100

0.0

Gannel

Gannel

1.0

300
200

0.5

Cu conc. (g g1)

Cu conc. (g g1)

100

Restronguet Creek
300
200
100
0

Tavy

0.0

Restronguet Creek
1.0
0.5
0.0

Tavy

1.0

300

0.5

200
100

0.0

Fishfood

Fishfood

1.0

300
200
100
0

0.5
0.0
0

10

15

20

Time (d)

10

15

20

Time (d)

Fig. 6. Palaemonetes varians. Relationships between Cu concentrations of predator decapods P. varians and number of
days feeding on diet of the polychaete worm Nereis diversicolor from 4 sites or on fishfood. Best-fit linear regression lines
are shown when significant (regression details in Table 6)

Fig. 7. Palaemonetes varians. Relationships between Cd concentrations of predator decapods P. varians and number of
days feeding on diet of the polychaete worm Nereis diversicolor from 4 sites or on fishfood. Best-fit linear regression lines
are shown when significant (regression details in Table 6)

cant decrease in the accumulated Ag concentration

(but no significant change in Fe or As concentrations)
of decapods fed the fishfood diet (Table 6). There was
no significant accumulation of Mn by P. varians from
N. diversicolor from 3 of the 4 sites, but a significant
decrease in the Mn concentration of decapods fed
Gannel worms (Table 6). There was also a significant
decrease in the accumulated Mn concentration of
decapods fed the fishfood diet (Table 6).

decapods on each of the 4 worm diets and the prey

concentration of Cu or Zn, or of added Cu + Zn, or
added Cu + Zn + Pb (all expressed in molar concentrations).

Mortality of predator decapod Palaemonetes varians

There was significant mortality of the decapods in
the experiment. Table 7 presents LT50 (time estimated
for 50% mortality) data calculated from the mortality
observed. As can be seen, survival on the fishfood diet
was excellent, while all Nereis diversicolor diets
caused mortality of Palaemonetes varians to varying
degrees. There were no significant relationships (bestfit linear regressions, p > 0.05) between the LT50 of the

DISCUSSION
The first questions we asked were whether the
predator Nereis virens accumulated trace metals from
a diet of its fellow polychaete worm N. diversicolor
from different metal-rich estuaries and, if so, whether
there was any effect of variable metal concentrations
in the diet.
Nereis virens clearly accumulated Cu from the diet,
with more Cu being accumulated from the more Curich diets across the range of diet concentrations
(Fig. 2, Table 4). The accumulation of Cu from the most
Cu-rich diet (Restronguet Creek N. diversicolor) confirmed the result of Rainbow et al. (2004), but significant Cu accumulation also occurred at lower dietary

177

Rainbow et al.: Trophic transfer of trace metals

Table 6. Palaemonetes varians. Regression details for relationships between

trace metal concentration (g g1) of predator decapods (y) and number of days
(x) feeding on a diet of the polychaete Nereis diversicolor from 1 of 4 sites or on
fishfood (TetraFin). Regression details are also given for relationships between
dry weight (g) of decapods (y) and number of days feeding (x). ns: not significant
Metal
Site/weight

Slope
SE

Intercept
SE

R2

Table 7. Palaemonetes varians. Estimated

time (d SD) to mortality of 50% of population (LT50) of the decapod crustacean fed a
diet of the polychaete worm Nereis diversicolor from 1 of 4 estuaries in SW England,
or of fishfood. LT50s of diets with different
indices differ significantly (p < 0.05)
Diet
Site

Copper
Dart
Gannel
Restronguet Creek
Tavy
Fishfood

0.349 1.290
0.529 1.369
2.737 1.468
3.149 2.571
5.922 1.096

162 10.0
158 10.9
156 11.7
154 14.1
165 13.4

0.003
0.006
0.127
0.086
0.519

25 0.789 ns
25 0.703 ns
26 0.074 ns
18 0.238 ns
29 < 0.001***

Zinc
Dart
Gannel
Restronguet Creek
Tavy
Fishfood

0.413 0.567
0.079 0.507
0.544 0.681
2.541 1.163
2.712 0.535

73.8 4.50
68.4 4.12
69.5 5.53
68.7 6.58
79.8 6.53

0.024
0.001
0.027
0.241
0.487

24 0.474 ns
24 0.877 ns
25 0.432 ns
17 0.045*
29 < 0.001***

Lead
Dart
Gannel
Restronguet Creek

0.016 0.144
0.027 0.037
0.386 0.766

Cadmium
Dart
Gannel
Restronguet Creek
Tavy
Fishfood

2.18 1.15 0.002

1.50 0.389 0.100
7.75 7.21 0.018

9
7
16

0.020 0.006
0.014 0.005
0.037 0.007
0.011 0.007
0.008 0.005

0.26 0.05
0.24 0.04
0.28 0.06
0.23 0.04
0.32 0.06

0.357
0.307
0.574
0.200
0.144

21 0.004**
22 0.007**
22 < 0.001***
13 0.125 ns
19 0.110 ns

Iron
Dart
Gannel
Restronguet Creek
Tavy
Fishfood

0.573 1.136
0.238 0.787
1.374 0.860
0.279 1.058
0.628 0.522

37.4 9.21
37.0 6.53
34.6 7.13
31.9 5.76
32.3 6.73

0.012
0.004
0.104
0.005
0.057

23
23
24
15
26

Silver
Dart
Gannel
Restronguet Creek
Tavy
Fishfood

0.003 0.008
0.016 0.014
0.012 0.008
0.004 0.011
0.034 0.007

0.92 0.07
0.93 0.11
0.90 0.06
0.91 0.06
0.94 0.09

0.004
0.058
0.098
0.007
0.467

25 0.753 ns
25 0.246 ns
26 0.120 ns
17 0.746 ns
29 < 0.001***

Arsenic
Dart
Gannel
Restronguet Creek
Tavy
Fishfood

0.048 0.028
0.035 0.043
0.057 0.056
0.011 0.065
0.162 0.089

6.66 0.21
6.41 0.35
6.41 0.45
6.69 0.34
6.66 0.21

0.132
0.030
0.044
0.002
0.132

22
24
25
15
18

Manganese
Dart
Gannel
Restronguet Creek
Tavy
Fishfood

0.109 0.067
0.146 0.063
0.110 0.067
0.194 0.118
0.254 0.033

4.68 0.53
4.55 0.51
4.53 0.55
4.69 0.62
5.62 0.40

0.106
0.197
0.107
0.162
0.688

24 0.120 ns
24 0.030*
24 0.118 ns
16 0.122 ns
29 < 0.001***

0.036 0.005
0.035 0.005
0.036 0.004
0.037 0.005
0.037 0.007

0.012
0.065
0.002
0.001
0.019

25
25
26
18
31

Weight
Dart
0.0003 0.0006
Gannel
0.0008 0.0006
Restronguet Creek 0.0001 0.0004
Tavy
0.0001 0.0009
Fishfood
0.0004 0.0005

0.914 ns
0.489 ns
0.622 ns

0.619 ns
0.765 ns
0.124 ns
0.796 ns
0.241 ns

0.096 ns
0.415 ns
0.315 ns
0.865 ns
0.087 ns

0.606 ns
0.217 ns
0.849 ns
0.921 ns
0.458 ns

Nereis diversicolor
Tavy
Dart
Restronguet Creek
Gannel
Fishfood

LT50 (d)

10.3A 1.0
16.0B 2.2
23.3C 4.1
27.2D 5.4
>1000

Cu concentrations. Accumulation over

the time period of the experiment
depends both on the initial assimilation efficiency (AE) and on the subsequent physiological handling of the
metal taken up, so defining the accumulation pattern of the organism for
that metal (Rainbow 2002). Clearly
from these results (and from the literature on N. diversicolor: see Introduction) N. virens (and by extension other
species of Nereis) is a strong net accumulator of Cu, with low excretion (or
possibly even no excretion: see Geffard et al. 2005) of Cu taken up.
Although more Cu was accumulated
from the most Cu-rich diets, the efficiency of trophic transfer (as measured
by a trophic transfer coefficient describing the proportion of ingested
metal retained by the predator)
decreased with increasing Cu concentration in the diet (Fig. 5). There is no
reason to believe that the accumulation pattern of Nereis virens for Cu
will differ (for example by changes in
relative Cu efflux rate) in relation to
different dietary intakes of Cu, so it is
likely that it was the Cu assimilation
efficiency that varied with dietary Cu
concentration. A probable explanation
of the relationship observed is that the
subcellular fractional distribution of
accumulated Cu in the prey N. diversicolor changes with increased accumulated Cu concentration, i.e. more Cu is
bound into insoluble detoxification
products with relatively low trophic
availability as measured by AE (Wal-

178

Mar Ecol Prog Ser 321: 167181, 2006

lace & Luoma 2003, Rainbow et al. 2006). Indeed,

Berthet et al. (2003) have shown that in N. diversicolor
from Restronguet Creek, 91%/9% of accumulated Cu
is in insoluble/soluble form, as opposed to 70%/30% in
Blackwater worms. Worms from both sites contain
most of the insoluble Cu in sulphur-rich Cu-containing
extracellular granules in the epicuticle (probably
derived from the breakdown of Cu-bearing metallothionein in lysosomes in epidermal cells), and these
granules were much more abundant in the Restronguet Creek worms (Mouneyrac et al. 2003).
As in the case of Cu, Nereis virens showed net accumulation of Zn from the diet of N. diversicolor from different metal-rich estuaries (Fig. 3, Table 4). Therefore,
N. virens did not regulate body concentrations of Zn
when Zn was taken up from the diet (at least over the
range of dietary Zn concentrations [Table 2] and time
scale used here), in contrast to the observed regulation
of Zn body concentrations by N. diversicolor (at least
when Zn is supplied in dissolved form in the laboratory
[Amiard et al. 1987], or as deduced from relatively constant body Zn concentrations in worms from estuaries
with a wide range of sediment Zn concentrations
[Bryan 1974, 1976]). Again as for Cu, there was a significant inverse relationship between the trophic transfer coefficient and the Zn concentration in the prey
(Fig. 5), although in this case (in the absence of disproportionately high dietary Zn concentrations) a linear
regression showed a significant fit to the data
(Table 5). The same explanation probably holds here,
i.e. the prey worms with the highest Zn concentrations
would have an increased proportion of the accumulated metal in insoluble and therefore less trophically
available form for assimilation. Again, Berthet et al.
(2003) have shown that 81%/19% of the accumulated
Zn in Restronguet Creek N. diversicolor is insoluble/soluble compared to 68%/32% in Blackwater
worms. The insoluble Zn is in spherocrystals in the
cells of the gut wall of Restronguet Creek worms, representing the major store of detoxified Zn in these
worms; the same spherocrystals are present in Blackwater worms, but do not contain Zn at detectable levels (Mouneyrac et al. 2003). Furthermore, Restronguet
Creek N. diversicolor exposed to radiolabelled Zn in
sediment accumulated a higher percentage of the new
labelled Zn in the metal-rich granule fraction than did
Blackwater worms, in correlation with a significantly
lower Zn AE in P. varians feeding on the Restronguet
Creek worms (Rainbow et al. 2006).
The same observations hold for Pb and Cd despite
limited data sets being available. Nereis virens showed
significant net Pb accumulation for 4 out of 5 diets of
N. diversicolor (Fig. 4, Table 4), and again there was a
significant inverse linear relationship between the
trophic transfer coefficient and prey Pb concentration

(Fig. 5, Table 5). There are few data on the subcellular

distribution of accumulated Pb in N. diversicolor.
Bryan (1976) made brief reference in a table to Pb in
epidermal granules in high-metal N. diversicolor, but
the reference quoted (Bryan 1974) contained no such
report. Nevertheless, it remains likely that increased
accumulated Pb concentrations in N. diversicolor are
associated with a greater degree of detoxification into
insoluble forms and subsequently reduced relative
trophic availability for assimilation by predators.
Cadmium concentrations in the Nereis diversicolor
prey did not vary much (Table 2) and did not include
greatly raised values (Bryan & Hummerstone 1973a,
Bryan et al. 1985). Nevertheless, N. virens feeding on
Blackwater prey did show significant accumulation of
Cd (Table 4), and there was a significant inverse linear
relationship between the trophic transfer coefficient
and prey Cd concentration (Fig. 5, Table 5). In agreement with the absence of striking differences in Cd
concentrations between prey types, Rainbow et al.
(2006) found AEs of newly accumulated Cd in Palaemonetes varians feeding on Restronguet Creek and
Blackwater N. diversicolor to be very similar, and there
were no differences between prey worms in the percentages of newly accumulated labelled Cd stored as
metal-rich granules. On the other hand, Berthet et al.
(2003) reported contradictory data, for they found
51%/49% of the accumulated Cd in Restronguet
Creek worms to be insoluble/soluble compared to
72%/28% in Blackwater worms.
For the most part, Nereis virens did not show net
accumulation of Fe from the diet (Table 4), and there
was no significant relationship between prey Fe concentrations and (albeit insignificant) trophic transfer
coefficients (Fig. 5, Table 5).
The second part of our study involved the decapod
crustacean Palaemonetes varians as the predator. Although we expected no increased accumulation of
either Cu or Zn, for which body concentrations are regulated in this decapod, it was expected that there
might be observable net accumulation of non-essential
metals such as cadmium and lead. In the event, the
belated availability of ICP metal analysis enabled measurements of a wider range of metals than was feasible
for the experiments involving Nereis virens. The same
principal question is relevant. Does the predator P. varians accumulate trace metals from a diet of the polychaete worm N. diversicolor from different metal-rich
estuaries?
As expected, Palaemonetes varians showed no net
accumulation of either Cu or Zn when feeding on diets
of N. diversicolor from 4 southwest England estuaries
(Fig. 6, Table 6). P. varians will certainly assimilate Zn
very efficiently from such a diet (41 to 84% AE: Rainbow et al. 2006), and so the newly assimilated Zn is not

Rainbow et al.: Trophic transfer of trace metals

being added to the existing body Zn as the decapods

regulate the Zn body concentration. The same process
presumably occurs for Cu (see Introduction), the absence of a suitable Cu radiotracer limiting any measurement of Cu AE.
There was also no significant accumulation of lead
by Palaemonetes varians from the diet of Nereis diversicolor from 3 estuaries (Table 6), despite prey Pb concentrations (Table 2) that are apparently high, at least
in the Gannel (Bryan et al. 1985). Decapod crustaceans
do not regulate body concentrations of Pb (Amiard et
al. 1987, Rainbow 1998), so some net accumulation
might have been expected. Although the kinetics of
the bioaccumulation of Pb can be followed using the
radioisotope 210Pb (for example Fisher et al. 1996), the
beta-emitting properties of this isotope do restrict its
suitability for the counting of live animals. In practice,
therefore, again the lack of a suitable radiotracer limits
comment on Pb assimilation efficiencies.
In spite of the fact that Cd concentrations in the
Nereis diversicolor prey (Table 2) are not particularly
high (Bryan & Hummerstone 1973a, Bryan et al. 1985),
Palaemonetes varians did show significant net accumulation of Cd for 3 out of 4 N. diversicolor diets
(Fig. 7, Table 6). Furthermore, in spite of the limited
range of prey Cd concentrations (Table 2), there was a
positive relationship (albeit insignificant) between
prey concentration and rate of Cd accumulation by
P. varians.
As for Nereis virens, there was no significant net
accumulation of Fe from the N. diversicolor diets by
Palaemonetes varians (Table 6), a result which may be
a feature of regulation of body Fe concentration by the
decapod given that Fe is an essential trace metal (Rainbow 1998). P. varians also showed no net accumulation
of Ag from the N. diversicolor diets (Table 6), perhaps
reflecting the relative lack of variation in Ag concentration between prey (Table 2) and the low total intake
of Ag from each diet. It cannot be concluded that the
assimilation efficiency of Ag by P. varians is particularly low, given the observed Ag assimilation efficiencies of this predator (10 to 50%) from radiolabelled
N. diversicolor prey (Rainbow et al. 2006). Similarly,
there was no net accumulation of As from the N. diversicolor diets by P. varians (Table 6), even in the case of
the Restronguet Creek prey worms with greatly increased accumulated As concentrations (Table 2). It is
not known whether palaemonid decapods can regulate body concentrations of As, but the evidence here
suggests that P. varians excreted any excess As load
taken up from a diet of Restronguet Creek N. diversicolor. There were no cases of significant net accumulation of Mn by P. varians from the N. diversicolor diets,
and there was even one case (Gannel) of a significant
decrease in Mn concentration in the predator (Table 6).

179

Given the large number of statistical tests carried out

in this study, this atypical result may have occurred by
chance.
A surprising feature of the experiment in which
Palaemonetes varians was fed Nereis diversicolor was
the mortality of the decapods on every worm diet. It
was not the experimental set up that caused mortality,
since there were significant differences in the LT50s of
the N. diversicolor diets (Table 7), and there was
almost no mortality when fishfood was provided as the
diet under the same experimental conditions. The
observed significant differences between the mortalities caused by the different worm diets indicate that
the cause of mortality was not a simple lack of an
essential ingredient in the worms as opposed to the
fishfood. The conclusion is that the differential mortality of the decapods was caused by differences between
the N. diversicolor prey. The catchments of the rivers
leading to the estuaries concerned typically lack industrial sources of pollution or large towns, and there are
no obvious differential sources of hydrocarbon or
organochlorine pollution. The prey worms clearly had
raised toxic metal concentrations (except perhaps in
the case of the Dart). These had not proved lethal to
the other predator N. virens, but the feeding rates of
the worms were transparently lower than those of the
decapod. Perhaps the cocktail of toxic metals in the
N. diversicolor diet proved lethal. The order of the
LT50s (Tavy < Dart < Restronguet Creek < Gannel),
particularly the high value for the Dart, is not helpful in
identifying a particular metal from a comparison of
metal concentrations in the worms (Table 2), and
indeed there was no significant relationship between
LT50 and molar concentrations of Cu, Zn, Cu + Zn, or
Cu + Zn + Pb. This lack of relationship would not be
unexpected if the increase in metal concentration in
the worm prey is offset by the increasing lack of
bioavailability of accumulated metal in the prey. In
short, it is difficult to come to a firm conclusion as to the
cause of mortality.
The use of fishfood as a diet for Palaemonetes varians
gave rise to an anomaly in the patterns of net accumulation of trace metals by the decapod. The decapods
fed on fishfood showed a significant net loss of accumulated Cu, Zn, Ag and Mn over the 18 d experiment,
but not of Cd, Fe or As. The reduction in metal concentration could not be attributed to a gain in weight
under this particular diet regime, for there was no significant change in weight of the decapods over the
course of the experiment, with this or any other diet
(Table 6). There is no clear reason for the observed
drop in accumulated concentrations. P. varians will
apparently assimilate metals efficiently from the fishfood (authors unpubl. obs.). Perhaps excretion of particular metals had been stimulated in some way by

180

Mar Ecol Prog Ser 321: 167181, 2006

physiological processes that resulted from this apparently good food supply. It is possible, for example, that
the fishfood diet increased the turnover of the epithelial cell cycle in the hepatopancreas of the decapods
(Gibson & Barker 1979), promoting the release of existing detoxified stores of metals in the epithelial cells
(Rainbow 1998).
Trophic transfer of trace metals from prey to predator
can be divided into 2 separate processes assimilation
and subsequent accumulation by the predator. The
first stage, assimilation, is controlled by the assimilation efficiency (Wang & Fisher 1999) of the host for the
particular metal from a particular food source. AE
depends on characteristics of the predators digestive
processes (presence and concentration of digestive
enzymes, gut pH, etc.) and on characteristics of the
metal in the prey item (concentration of metal, subcellular fractionation and associated form of chemical
binding of the metal, etc.). The second stage, accumulation, depends on the accumulation pattern of the
predator for the particular metal, accumulation patterns varying between predators (often at quite low
taxonomic level) and intraspecifically between metals
(Rainbow 2002).
The results obtained here show that trace metals
accumulated in the polychaete worm Nereis diversicolor can be assimilated (and potentially subsequently
accumulated) by predators with the potential for further transport along estuarine food chains to top predators such as fishes or birds (Rainbow et al. 2004). Accumulation of metals from the prey by the predator
N. virens did typically increase with increasing metal
concentration in the prey, although the coefficient of
trophic transfer typically decreased with increased
prey concentration, probably via an effect at the assimilation stage controlled by the subcellular distribution
of metals in the prey into components of different relative availabilities for assimilation (Wallace & Luoma
2003, Rainbow et al. 2006).
In the case of the decapod Palaemonetes varians as a
predator, the trace metal accumulation patterns of this
palaemonid decapod (for example regulation of body
concentrations of essential metals) appear to act as a
brake for trophic transfer of metals from prey, subsequent to what appears to be efficient assimilation from
the Nereis diversicolor prey (Rainbow et al. 2006). The
toxicity of a trace metal to a recipient animal is not necessarily related to its accumulated concentration
much of the accumulated metal being in detoxified
form (Rainbow 2002). Rather, it is the rate of trace
metal uptake (from all sources typically from solution
and diet together) that controls toxicity; when the integrated rate of uptake exceeds the combined rates of
excretion and detoxification by the animal concerned,
then the concentration of metabolically available metal

in the body increases and toxic effects occur (Rainbow

2002). It is possible that, in this study, P. varians experienced an uptake of the cocktail of metals present in
the diet of N. diversicolor with raised, accumulated,
toxic metal concentrations that was sufficient to achieve
toxicity without evidence of increased accumulation of
assimilated metals to atypically high body concentrations.
As shown here in the specific case of the polychaete
worm Nereis diversicolor, metal-rich invertebrates that
have accumulated metals from the rich historical store
in the sediments of particular estuaries in southwest
England can potentially pass these metals along food
chains, accumulation and total food chain transfer
depending on the metal assimilation efficiencies and
accumulation patterns of the animal at each trophic
level. This trophic transfer may be significant enough
to have ecotoxicological effects.
Acknowledgements. This research was partly supported by
funding from the International Lead Zinc Research Organization and Rio Tinto. S. N. Luomas contribution was partly
funded by a W.J. Fulbright Distinguished Scholar Award. L.P.
was supported by a grant from the CNRS, France (PICS 1056).
We are grateful to J. C. Amiard for analysis of mortality data
and Jeanne Dileo for artwork. We thank Professor N. S. Fisher
for constructive criticism of the manuscript.

LITERATURE CITED
Amiard JC, Amiard-Triquet C, Berthet B, Mtayer C (1987)
Comparative study of the patterns of bioaccumulation of
essential (Cu,Zn) and non-essential (Cd, Pb) trace metals
in various estuarine and coastal organisms. J Exp Mar Biol
Ecol 106:7389
Berthet B, Mouneyrac C, Amiard JC, Amiard-Triquet C,
Beaunier P, Jeantet AY, Smith BD, Rainbow PS (2003)
Responses to metals of the polychaete annelid Hediste
diversicolor, a key species in estuarine and coastal sediments. Arch Environ Contam Toxicol 45:468478
Blackmore G (2000) Field evidence of metal transfer from
invertebrate prey to an intertidal predator, Thais clavigera
(Gastropoda: Muricidae). Estuar Coast Shelf Sci 51:
127139
Blackmore G, Morton B (2002) The influence of diet on comparative trace metal cadmium, copper and zinc accumulation in Thais clavigera (Gastropoda: Muricidae) preying
on intertidal barnacles or mussels. Mar Pollut Bull 44:
870876
Bryan GW (1968) Concentrations of zinc and copper in the tissues of decapod crustaceans. J Mar Biol Assoc UK 48:
303321
Bryan GW (1974) Adaptation of an estuarine polychaete to
sediments containing high concentrations of heavy metals. In: Vernberg FJ, Vernberg WB (eds) Pollution and
physiology of marine organisms. Academic Press, New
York, p 123135
Bryan GW (1976) Some aspects of heavy metal tolerance in
aquatic organisms. In: Lockwood APM (ed) Effects of pollutants on aquatic organisms. Cambridge University Press,
Cambridge, p 734

Rainbow et al.: Trophic transfer of trace metals

181

Bryan GW, Gibbs PE (1980) Metals in nereid polychaetes: the

contribution of metals in the jaws to the total body burden.
J Mar Biol Assoc UK 60:641654
Bryan GW, Gibbs PE (1983) Heavy metals in the Fal estuary,
Cornwall: a study of long-term contamination by mining
waste and its effects on estuarine organisms. Occas Publ
Mar Biol Assoc UK 2:1112
Bryan GW, Hummerstone LG (1971) Adaptation of the polychaete Nereis diversicolor to estuarine sediments containing high concentrations of heavy metals. I. General observations and adaptation to copper. J Mar Biol Assoc UK 51:
845863
Bryan GW, Hummerstone LG (1973a) Adaptation of the polychaete Nereis diversicolor to estuarine sediments containing high concentrations of zinc and cadmium. J Mar Biol
Assoc UK 53:839857
Bryan GW, Hummerstone LG (1973b) Adaptation of the polychaete Nereis diversicolor to manganese in estuarine sediments. J Mar Biol Assoc UK 53:859872
Bryan GW, Langston WJ, Hummerstone LG (1980) The use of
biological indicators of heavy metal contamination in estuaries. Occas Publ Mar Biol Assoc UK 1:173
Bryan GW, Langston WJ, Hummerstone, Burt GR (1985) A
guide to the assessment of heavy metal contamination in
estuaries using biological indicators. Occas Publ Mar Biol
Assoc UK 4:192
Cheung MS, Wang WX (2005) Influence of subcellular compartmentalization in different prey on the transfer of metals to a predatory gastropod. Mar Ecol Prog Ser 286:
155166
Chong K, Wang WX (2000) Assimilation of Cd, Cr and Zn by
the green mussel Perna viridis and the clam Ruditapes
philippinarum. Environ Toxicol Chem 19:16601667
Croteau MN, Luoma SN, Topping BR, Lopez CB (2004) Stable
metal isotopes reveal copper accumulation and loss
dynamics in the freshwater bivalve Corbicula. Environ Sci
Technol 38:502509
Dines HG (1969) The metalliferous mining region of SouthWest England. Her Majestys Stationery Office, London
Fisher NS, Teyssi JL, Fowler SW, Wang WX (1996). Accumulation and retention of metals in mussels from food and
water: a comparison under field and laboratory conditions.
Environ Sci Technol 30:32323242
Geffard A, Smith BD, Amiard-Triquet C, Jeantet AY, Rainbow
PS (2005) Kinetics of trace metal accumulation and excretion in the polychaete Nereis diversicolor. Mar Biol
147:12911304
Gibson R, Barker PL (1979) The decapod hepatopancreas.
Oceanogr Mar Biol Annu Rev 17:285346
Luoma SN, Rainbow PS (2005) Why is metal bioaccumulation
so variable? Biodynamics as a unifying concept. Environ
Sci Technol 39:19211931
Mouneyrac C, Mastain O, Amiard JC, Amiard-Triquet C,
Beaunier P, Jeantet AY, Smith BD, Rainbow PS (2003)
Trace-metal detoxification and tolerance of the estuarine
worm Hediste diversicolor chronically exposed in their
environment. Mar Biol 143:731744
Ng TYT, Amiard-Triquet C, Rainbow PS, Amiard JC, Wang
WX (2005) Physico-chemical form of trace metals accumulated by phytoplankton and their assimilation by filterfeeding invertebrates. Mar Ecol Prog Ser 299:179191
Nugegoda D, Rainbow PS (1989) Effects of salinity changes
on zinc uptake and regulation by the decapod crustaceans
Palaemon elegans and Palaemonetes varians. Mar Ecol
Prog Ser 51:5775
Rainbow PS (1998) Phylogeny of trace metal accumulation in
crustaceans. In: Langston WJ, Bebianno M (eds) Metal

metabolism in aquatic environments. Chapman & Hall,

London, p 285319
Rainbow PS (2002) Trace metal concentrations in aquatic
invertebrates: why and so what? Environ Pollut 120:
497507
Rainbow PS, Blackmore G, Wang WX (2003) Effects of previous-field exposure history on the uptake of trace metals
from water and food by the barnacle Balanus amphitrite.
Mar Ecol Prog Ser 259:201213
Rainbow PS, Geffard A, Jeantet AY, Smith BD, Amiard JC,
Amiard-Triquet C (2004) Enhanced food-chain transfer of
copper from a diet of copper-tolerant estuarine worms.
Mar Ecol Prog Ser 271:183191
Rainbow PS, Poirier L, Smith BD, Brix KV, Luoma SN (2006)
Trophic transfer of trace metals: subcellular compartmentalization in a polychaete and assimilation by a decapod
crustacean. Mar Ecol Prog Ser 308:91100
Reinfelder JR, Fisher NS (1991) The assimilation of elements
ingested by marine copepods. Science 251:794796
Reinfelder JR, Wang WX, Luoma SN, Fisher NS (1997) Assimilation efficiencies and turnover rates of trace elements in
marine bivalves: a comparison of oysters, clams and mussels. Mar Biol 129:443452
Wallace WG, Lopez GR (1996) Relationship between the subcellular cadmium distribution in prey and cadmium transfer to a predator. Estuaries 19:923930
Wallace WG, Lopez GR (1997) Bioavailability of biologically
sequestered cadmium and the implications of metal detoxification. Mar Ecol Prog Ser 147:149157
Wallace WG, Luoma SN (2003) Subcellular compartmentalization of Cd and Zn in two bivalves. II. The significance of
trophically available metal (TAM). Mar Ecol Prog Ser 257:
125137
Wang WX (2002) Interactions of trace metals and different
marine food chains. Mar Ecol Prog Ser 243:295309
Wang WX, Fisher NS (1996) Assimilation of trace elements
and carbon by the mussel Mytilus edulis: effects of food
composition. Limnol Oceanogr 41:197207
Wang WX, Fisher NS (1999) Assimilation efficiencies of
chemical contaminants in aquatic invertebrates: a synthesis. Environ Toxicol Chem 18:20342045
Wang WX, Rainbow PS (2000) Dietary uptake of Cd, Cr, and
Zn by the barnacle Balanus trigonus: influence of diet
composition. Mar Ecol Prog Ser 204:159168
White SL, Rainbow PS (1982) Regulation and accumulation of
copper, zinc and cadmium by the shrimp Palaemon elegans. Mar Ecol Prog Ser 8:95101
White SL, Rainbow PS (1984a) Regulation of zinc concentration by Palaemon elegans (Crustacea: Decapoda): zinc
flux and effects of temperature, zinc concentration and
moulting. Mar Ecol Prog Ser 16:135147
White SL, Rainbow PS (1984b) Zinc flux in Palaemon elegans
(Crustacea: Decapoda): moulting, individual variation and
tissue distribution. Mar Ecol Prog Ser 19:153166
White SL, Rainbow PS (1986) Accumulation of cadmium by
Palaemon elegans (Crustacea: Decapoda). Mar Ecol Prog
Ser 32:1725
Xu Y, Wang WX (2002) The assimilation of detritus-bound
metals by the marine copepod Acartia spinicauda. Limnol
Oceanogr 47:604610
Xu Y, Wang WX (2004) Silver uptake by a marine diatom and
its transfer to the coastal copepod Acartia spinicauda.
Environ Toxicol Chem 23:682690
Zhou Q, Rainbow PS, Smith BD (2003) Tolerance and accumulation of the trace metals zinc, copper and cadmium in
three populations of the polychaete Nereis diversicolor.
J Mar Biol Assoc UK 83:6572

Editorial responsibility: Howard I. Browman (Associate

Editor-in-Chief), Storeb, Norway

Submitted: October 6, 2005; Accepted: January 15, 2006

Proofs received from author(s): August 21, 2006